A Novel Challenge Set for Hebrew Morphological Disambiguation and Diacritics Restoration

TL;DR

This paper introduces a new challenge set for Hebrew morphological disambiguation and diacritics restoration, addressing unbalanced ambiguity issues and significantly improving state-of-the-art performance.

Contribution

It provides the first challenge dataset for Hebrew homographs with balanced analysis representation, enabling better evaluation and training of disambiguation models.

Findings

Current SOTA performs poorly on unbalanced ambiguities

New dataset improves F1 score from 0.67 to 0.95

Annotated datasets are publicly available

Abstract

One of the primary tasks of morphological parsers is the disambiguation of homographs. Particularly difficult are cases of unbalanced ambiguity, where one of the possible analyses is far more frequent than the others. In such cases, there may not exist sufficient examples of the minority analyses in order to properly evaluate performance, nor to train effective classifiers. In this paper we address the issue of unbalanced morphological ambiguities in Hebrew. We offer a challenge set for Hebrew homographs -- the first of its kind -- containing substantial attestation of each analysis of 21 Hebrew homographs. We show that the current SOTA of Hebrew disambiguation performs poorly on cases of unbalanced ambiguity. Leveraging our new dataset, we achieve a new state-of-the-art for all 21 words, improving the overall average F1 score from 0.67 to 0.95. Our resulting annotated datasets are made publicly available for further research.